#dos

Xenstore: Guests can cause Xenstore to not free temporary memory When working on a request of a guest, xenstored might need to allocate quite large amounts of memory temporarily. This memory is freed only after the request has been finished completely. A request is regarded to be finished only after the guest has read the response message of the request from the ring page. Thus a guest not reading the response can cause xenstored to not free the temporary memory. This can result in memory shortages causing Denial of Service (DoS) of xenstored.

Xenstore: Cooperating guests can create arbitrary numbers of nodes T[his CNA information record relates to multiple CVEs; the text explains which aspects/vulnerabilities correspond to which CVE.] Since the fix of XSA-322 any Xenstore node owned by a removed domain will be modified to be owned by Dom0. This will allow two malicious guests working together to create an arbitrary number of Xenstore nodes. This is possible by domain A letting domain B write into domain A's local Xenstore tree. Domain B can then create many nodes and reboot. The nodes created by domain B will now be owned by Dom0. By repeating this process over and over again an arbitrary number of nodes can be created, as Dom0's number of nodes isn't limited by Xenstore quota.

Xenstore: Guests can get access to Xenstore nodes of deleted domains Access rights of Xenstore nodes are per domid. When a domain is gone, there might be Xenstore nodes left with access rights containing the domid of the removed domain. This is normally no problem, as those access right entries will be corrected when such a node is written later. There is a small time window when a new domain is created, where the access rights of a past domain with the same domid as the new one will be regarded to be still valid, leading to the new domain being able to get access to a node which was meant to be accessible by the removed domain. For this to happen another domain needs to write the node before the newly created domain is being introduced to Xenstore by dom0.

Xenstore: guests can let run xenstored out of memory T[his CNA information record relates to multiple CVEs; the text explains which aspects/vulnerabilities correspond to which CVE.] Malicious guests can cause xenstored to allocate vast amounts of memory, eventually resulting in a Denial of Service (DoS) of xenstored. There are multiple ways how guests can cause large memory allocations in xenstored: - - by issuing new requests to xenstored without reading the responses, causing the responses to be buffered in memory - - by causing large number of watch events to be generated via setting up multiple xenstore watches and then e.g. deleting many xenstore nodes below the watched path - - by creating as many nodes as allowed with the maximum allowed size and path length in as many transactions as possible - - by accessing many nodes inside a transaction

### Impact The package muhammara before 2.6.0; all versions of package hummus are vulnerable to Denial of Service (DoS) when supplied with a maliciously crafted PDF file to be appended to another. ### Patches It has been patched in 2.6.0 for muhammara and not at all for hummus ### Workarounds Do not process files from untrusted sources ### References PR: https://github.com/julianhille/MuhammaraJS/pull/194 Issue: https://github.com/julianhille/MuhammaraJS/issues/191 Issue in hummus: https://github.com/galkahana/HummusJS/issues/293 ### Outline differences to https://nvd.nist.gov/vuln/detail/CVE-2022-25892 The difference is one is in [src/deps/PDFWriter/PDFParser.cpp](https://github.com/julianhille/MuhammaraJS/commit/1890fb555eaf171db79b73fdc3ea543bbd63c002#diff-09ac2c64aeab42b14b2ae7b11a5648314286986f8c8444a5b3739ba7203b1e9b) and the other is [PDFDocumentHandler.cpp](https://github.com/julianhille/MuhammaraJS/pull/194/files#diff-38d338ea4c047fd7dd9a05b5ffe7c964f0fa7e79aff4c307ccee75...

### Impact The package muhammara before 2.6.1, from 3.0.0 and before 3.1.1; all versions of package hummus are vulnerable to Denial of Service (DoS) when supplied with a maliciously crafted PDF file to be parsed. ### Patches It has been patched in 3.1.1 and has been backported to 2.6.1 There is no patch for hummus ### Workarounds Do not process files from untrusted sources or update. ### References https://nvd.nist.gov/vuln/detail/CVE-2022-25892 https://github.com/galkahana/HummusJS/issues/463 https://github.com/julianhille/MuhammaraJS/issues/214 https://github.com/julianhille/MuhammaraJS/commit/1890fb555eaf171db79b73fdc3ea543bbd63c002 https://github.com/julianhille/MuhammaraJS/commit/90b278d09f16062d93a4160ef0a54d449d739c51 https://security.snyk.io/vuln/SNYK-JS-HUMMUS-3091138 https://security.snyk.io/vuln/SNYK-JS-MUHAMMARA-3060320

The package muhammara before 2.6.0; all versions of package hummus are vulnerable to Denial of Service (DoS) when PDFStreamForResponse() is used with invalid data.

The package muhammara before 2.6.1, from 3.1.0 and before 3.1.1; all versions of package hummus are vulnerable to Denial of Service (DoS) when supplied with a maliciously crafted PDF file to be parsed.

conduit-hyper integrates a conduit application with the hyper server. Prior to version 0.4.2, `conduit-hyper` did not check any limit on a request's length before calling [`hyper::body::to_bytes`](https://docs.rs/hyper/latest/hyper/body/fn.to_bytes.html). An attacker could send a malicious request with an abnormally large `Content-Length`, which could lead to a panic if memory allocation failed for that request. In version 0.4.2, `conduit-hyper` sets an internal limit of 128 MiB per request, otherwise returning status 400 ("Bad Request"). This crate is part of the implementation of Rust's [crates.io](https://crates.io/), but that service is not affected due to its existing cloud infrastructure, which already drops such malicious requests. Even with the new limit in place, `conduit-hyper` is not recommended for production use, nor to directly serve the public Internet.

### Impact Attackers can create long chains of CAs that would lead to OctoRPKI exceeding its max iterations parameter that would cause the program to crash and not finish the validation and thus a denial of service. ### Patches This issue is fixed in v1.4.4 ### Workarounds None.